MACHINE LEARNING

VolverGo back

Resultados 98 results. LastUpdate Updated on 06/12/2021 [04:08:00] pdf PDF xls XLS

Solicitudes publicadas en los últimos 30 días / Applications published in the last 30 days



Page1/4 nextPage   results/page


METHODS AND SYSTEMS FOR DYNAMIC ADJUSTMENT OF A LANDING PAGE

Publication No.: US2021374827A1 02/12/2021

Applicant:

CAPITAL ONE SERVICES LLC [US]

Absstract of: US2021374827A1

A computer-implemented method for dynamically adjusting a landing page with a personalized recommendation to a user may include obtaining first image data of one or more vehicles via a device associated with the user; obtaining second image data of the one or more vehicles based on the first image data, wherein the second image data comprises at least a subset of the one or more images of the one or more vehicles; determining user preference data based on the second image data of the one or more vehicles via a trained machine learning algorithm, wherein the user preference data comprises one or more features of a user-preferred vehicle; determining the personalized recommendation to the user based on the user preference data, wherein the personalized recommendation comprises a personalized webpage showing information related to the user-preferred vehicle; and presenting, to the user, the personalized recommendation.

traducir

USING EDITOR SERVICE TO CONTROL ORCHESTRATION OF GRAMMAR CHECKER AND MACHINE LEARNED MECHANISM

Publication No.: US2021374340A1 02/12/2021

Applicant:

MICROSOFT TECHNOLOGY LICENSING LLC [US]

Absstract of: US2021374340A1

An editor service receives a textual input. The editor service provides the textual input to a rule-based grammar checker to obtain a grammar checker result. The editor service also provides the textual input to a machine learning (ML) fluency model that checks the textual input for errors and provides a ML model result. The editor service aggregates the grammar checker result and the ML model result and generates an editor service output based upon the aggregated results. A representation of the editor service result is provided to the client computing system for surfacing through a user interface.

traducir

POINTS OF INTEREST IN A LOCATION SHARING SYSTEM

Publication No.: US2021377693A1 02/12/2021

Applicant:

SNAP INC [US]

US_2020314586_A1

Absstract of: US2021377693A1

Systems, and methods for predicting that a user is located at a labeled place corresponding to a point of interest. A server computer accesses historical data comprising location data, and wireless network data collected from a plurality of client devices of a plurality of users over a period of time. For one or more labeled places, the data points corresponding to one of the users being located at the labeled place are identified. A labeled dataset is generated by tagging the identified data points with a label corresponding to the corresponding labeled place. A machine learning model is trained on the labeled dataset, so that when current location data are receiving from a client device of a user, it is possible to determine, using the trained machine learning model, whether the user is located at one of the one or more labeled places.

traducir

EARLY WARNING AND COLLISION AVOIDANCE

Publication No.: US2021375138A1 02/12/2021

Applicant:

DERQ INC [VG]

JP_2021518623_A

Absstract of: US2021375138A1

Among other things, equipment is located at an intersection of a transportation network. The equipment includes an input to receive data from a sensor oriented to monitor ground transportation entities at or near the intersection. A wireless communication device sends to a device of one of the ground transportation entities, a warning about a dangerous situation at or near the intersection, there is a processor and a storage for instructions executable by the processor to perform actions including the following. A machine learning model is stored that can predict behavior of ground transportation entities at or near the intersection at a current time. The machine learning model is based on training data about previous motion and related behavior of ground transportation entities at or near the intersection. Current motion data received from the sensor about ground transportation entities at or near the intersection is applied to the machine learning model to predict imminent behaviors of the ground transportation entities. An imminent dangerous situation for one or more of the ground transportation entities at or near the intersection is inferred from the predicted imminent behaviors. The wireless communication device sends the warning about the dangerous situation to the device of one of the ground transportation entities.

traducir

FACILITATING FRAUD DISPUTE RESOLUTION USING MACHINE LEARNING

Publication No.: US2021374764A1 02/12/2021

Applicant:

STATE FARM MUTUAL AUTOMOBILE INSURANCE CO [US]

US_2021374753_A1

Absstract of: US2021374764A1

In a computer-implemented method of facilitating a fraud dispute resolution process, types of information historically indicative of fraud (or its absence) may be identified by training a machine learning program using transaction data associated with financial transactions and fraud determinations for those transactions. An indication that fraud is suspected for a first transaction may be received, and transaction data may be retrieved. Based upon at least one of the identified types of information and the transaction data, a first set of one or more queries that are designed to ascertain whether the first transaction was fraudulent may be generated. The first set of queries may be transmitted to a remote computing device for display to the customer, and a first set of one or more customer responses may be received. Based upon the first set of customer responses, it may be determined whether the first transaction was fraudulent.

traducir

FEATURE REMOVAL FRAMEWORK TO STREAMLINE MACHINE LEARNING

Publication No.: US2021374562A1 02/12/2021

Applicant:

MICROSOFT TECHNOLOGY LICENSING LLC [US]

Absstract of: US2021374562A1

The disclosed embodiments provide a system for streamlining machine learning. During operation, the system determines a resource overhead for a baseline version of a machine learning model that uses a set of features to produce entity rankings and a number of features to be removed to lower the resource overhead to a target resource overhead. Next, the system calculates importance scores for the features, wherein each importance score represents an impact of a corresponding feature on the entity rankings. The system then identifies a first subset of the features to be removed as the number of features with lowest importance scores and trains a simplified version of the machine learning model using a second subset of the features that excludes the first subset of the features. Finally, the system executes the simplified version to produce new entity rankings.

traducir

GREEN ARTIFICIAL INTELLIGENCE IMPLEMENTATION

Publication No.: US2021374561A1 02/12/2021

Applicant:

BANK OF AMERICA [US]

Absstract of: US2021374561A1

A model designer creates models for machine learning applications while focusing on reducing the carbon footprint of the machine learning application. The model designer can automatically extract features of a machine learning application from requirements documents and automatically generate source code to implement that machine learning application. The model designer then uses computing statistics of previous models and machine learning applications to determine hardware limitations or restrictions to be placed on machine learning application or model. The designer then adds or adjusts the source code to enforce these hardware limitations and restrictions.

traducir

Machine learning to determine command and control sites

Publication No.: US2021377304A1 02/12/2021

Applicant:

ZSCALER INC [US]

Absstract of: US2021377304A1

Systems and methods include receiving a domain for a determination of a likelihood the domain is a command and control site; analyzing the domain with an ensemble of a plurality of trained machine learning models including a Uniform Resource Locator (URL) model that analyzes lexical features of a hostname of the domain and an artifact model that analyzes content features of a webpage associated with the domain; and combining results of the ensemble to predict the likelihood the domain is a command and control site.

traducir

METHOD AND SYSTEM FOR MACHINE LEARNING MODEL TESTING AND PREVENTIVE MEASURE RECOMMENDATION

Publication No.: US2021377286A1 02/12/2021

Applicant:

TATA CONSULTANCY SERVICES LTD [IN]

Absstract of: US2021377286A1

Data-driven applications depend on training data obtained from multiple internal and external data sources. Hence poisoning of the training data can cause adverse effects in the data driven applications. Conventional methods identifies contaminated test samples and avert them from entering into the training. A generic approach covering all data-driven applications and all types of data poisoning attacks in an efficient manner is challenging. Initially, data aggregation is performed after receiving a ML application for testing. A plurality of feature vectors are extracted from the aggregated data and a poisoned data set is generated. A plurality of personas are generated and are further prioritized to obtain a plurality of attack personas. Further, a plurality of security assessment vectors are computed for each of the plurality of attack personas. A plurality of preventive measures are recommended for each of the plurality of attack personas based on the corresponding security assessment vector.

traducir

UTILIZING WEB APPLICATON FIREWALL AND MACHINE LEARNING TO DETECT COMMAND AND CONTROL

Publication No.: US2021377295A1 02/12/2021

Applicant:

SAUDI ARABIAN OIL CO [SA]

Absstract of: US2021377295A1

A method for detecting Command and Control (C&C) toward a web application in a network includes: obtaining, using a Web Application Firewall (WAF) of the network, network traffic between the web application and a server outside the network; transmitting the network traffic from the WAF to a machine learning model; determining, using the machine learning model, whether the network traffic includes a command signature; in response to determining that the network traffic includes a command signature, generating a notification; and determining, based on the notification, whether the server is a C&C.

traducir

ALERT RULE EVALUATION FOR MONITORING OF LATE ARRIVING DATA

Publication No.: US2021374130A1 02/12/2021

Applicant:

MICROSOFT TECHNOLOGY LICENSING LLC [US]

Absstract of: US2021374130A1

A monitoring system is configured to distinguish between two types of alert rules—namely, invariant alert rules and variant alert rules—and to apply a different method of alert rule evaluation to each, wherein each alert rule evaluation method deals with the issue of latent data ingestion in a different way. By tailoring the alert rule evaluation method to the type of alert rule being evaluated, the system can apply an optimized approach for each type of alert rule in terms of achieving a trade-off between alert latency, alert accuracy, and cost of goods sold. In an embodiment, the system utilizes a machine learning model to classify a query associated with an alert rule as either increasing or non-increasing. Then, based on the query classification and a condition associated with the alert rule, the system determines if the alert rule is invariant or variant.

traducir

METHOD FOR REPRODUCIBILITY OF DEEP LEARNING CLASSIFIERS USING ENSEMBLES

Publication No.: US2021374500A1 02/12/2021

Applicant:

HITACHI LTD [JP]

Absstract of: US2021374500A1

Example implementations described herein involve systems and methods for generating an ensemble of deep learning or neural network models, which can involve, for a training set of data, generating a plurality of model samples for the training set of data, the plurality of model samples generated from deep learning or neural network methods; and aggregating output of the model samples to generate an output of the ensemble models.

traducir

METHOD AND SYSTEM FOR MONITORING A HEALTH OF A POWER CABLE ACCESSORY BASED ON MACHINE LEARNING

Publication No.: US2021373063A1 02/12/2021

Applicant:

3M INNOVATIVE PROPERTIES CO [US]

CN_112673265_A

Absstract of: US2021373063A1

Techniques, systems and articles are described for monitoring electrical equipment of a power grid and predicting likelihood failure events of such electrical equipment. In one example, a system includes an article of electrical equipment, at least one processor, and a storage device. The article of electrical equipment includes one or more sensors that are configured to generate sensor data indicative of one or more conditions of the article of electrical equipment. The storage device includes instructions that, when executed by the at least one processor, cause the at least one processor to: receive the sensor data; determine, based at least in part on the sensor data, a health of the article of electrical equipment; and responsive to determining the health of the article of electrical equipment, perform an operation.

traducir

SYSTEM AND METHOD FOR RECONSTRUCTING A 3D HUMAN BODY UNDER CLOTHING

Publication No.: US2021375045A1 02/12/2021

Applicant:

VIETTEL GROUP [VN]

Absstract of: US2021375045A1

The invention presents a system and a method for digitizing body shape from dressed human image using machine learning and optimization techniques. The invention is able to rapidly and accurately reconstruct human body shape without using costly, bulky and hazardous 3D scanners. Firstly, the system reconstructing human body shape from the dressed human image includes 2 main modules and 2 supplementary blocks, which are: (1) Input Block, (2) Pre-Processing Module, (3) Optimization Module, (4) Output Block. In which, the Pre-Processing Module comprises 4 blocks: (1) Image Standardization, (2) Clothes Classification and Segmentation, (3) Human Pose Estimation, (4) Cloth-Skin Displacement Model. The Optimization Modules comprises 2 blocks: (1) Human Parametric Model, (2) Human Parametric Optimization. Secondly, the method for reconstructing body shape from dressed human image includes 4 steps: (1) Collecting dressed human images, (2) Standardizing and extracting image information, (3) Parameterizing and optimizing human shape, (4) Displaying human body shape.

traducir

SYSTEM AND METHOD FOR AUTONOMOUS LEARNING OF CONTENTS USING A MACHINE LEARNING ALGORITHM

Publication No.: US2021375148A1 02/12/2021

Applicant:

FOUNTECH SOLUTIONS LTD [CY]

Absstract of: US2021375148A1

A method of autonomous learning of contents using a machine learning model, wherein in the method includes processing a corpus of data using the machine learning model to extract insights including at least one of a key knowledge, a topic of knowledge, a plurality of key entities, or an associated cognitive ability, collectively referred to as meta-tags; creating associations between a plurality of sub-components of at least one of an existing content, an enriched content, or one or more meta-tags, using the machine learning model to generate a graph knowledge base; and automatically performing at least one of: 1) building the graph knowledge base or 2) enriching an existing knowledge base, using the machine learning model and the one or more meta-tags, wherein the graph knowledge base comprises at least a graph form of information, wherein the graph knowledge base enables automatic retrieval of organized content to be used for generating teaching material for the user based on the one or more meta-tags.

traducir

Method and Apparatus for Teaching Using a Machine Learning Algorithm

Publication No.: US2021375151A1 02/12/2021

Applicant:

SOFFOS INC [US]

Absstract of: US2021375151A1

A method for teaching a student a selected content using a computer implemented machine learning algorithm. After the selected content has been presented, a query is presented to the student relating to that content. A correctness metric is developed as a function of the response of the student to the query. From the correctness metric, an inference is made of the comprehension by the student of that content. During this process, physiological indicia of the response of the student are used to develop a behavioral pattern. Using the pattern, additional content may be selected for presentation to the student. Over time, the machine learning algorithm is able iteratively to make improved inferences of the preferred learning method of the student as a function of the inferred comprehension of the student of each newly presented content.

traducir

SYSTEMS AND METHODS FOR DETERMINING AND USING HEALTH CONDITIONS BASED ON MACHINE LEARNING ALGORITHMS AND A SMART VITAL DEVICE

Publication No.: US2021375456A1 02/12/2021

Applicant:

AETNA INC [US]

Absstract of: US2021375456A1

In some instances, the disclosure provides a method performed by a smart vital device. The method comprises receiving sensor information indicating one or more health characteristics associated with an individual, wherein the sensor information comprises audio information indicating audio signals from a surrounding environment and temperature information indicating temperature readings from the surrounding environment, determining one or more health audio characteristics of the individual based on inputting the audio signals into one or more health condition machine learning datasets, determining one or more health temperature characteristics of the individual based on the temperature readings from the surrounding environment, determining one or more health conditions of the individual based on the one or more health audio characteristics and the one or more health temperature characteristics, and outputting the one or more health conditions of the individual.

traducir

METHOD AND APPARATUS FOR DETERMINING VEHICLE CLASS BASED UPON AUDIO DATA

Publication No.: US2021375305A1 02/12/2021

Applicant:

HERE GLOBAL BV [NL]

Absstract of: US2021375305A1

A method, apparatus and computer program product are provided to identify the class of vehicle driving over a road surface based upon audio data collected as the vehicle drives thereover. With respect to predicting a class of a vehicle, audio data is obtained that is created by the vehicle while driving over the road surface. The audio data includes one or more audio frequency features and/or one or more audio amplitude features. The audio data including the one or more audio frequency features and/or the one or more audio amplitude features is provided to a machine learning model and the class of the vehicle that created the audio data is predicted utilizing the machine learning model. A method, apparatus and computer program product are also provided for training the machine learning model to predict the class of the vehicle driving over the road surface.

traducir

METHOD OF AND SYSTEM FOR GENERATING A TRAINING SET FOR A MACHINE LEARNING ALGORITHM (MLA)

Publication No.: US2021374606A1 02/12/2021

Applicant:

YANDEX EUROPE AG [CH]

Absstract of: US2021374606A1

There is disclosed a method and system for generating a training set for training a machine learning algorithm (MLA) implemented in an information retrieval system. The method is executable by the server and comprises: retrieving, from a search log database of the server, a first query previously submitted to the server, a first SERP associated with the first query, a second query different from the first query and submitted after the first query, and a second SERP associated with the second query, the first query and the second query having been submitted by the electronic device: the first SERP including a first set of search results; and the second SERP including a second set of search results; in response to the second query being submitted within a same search session as the first query, generating the training set to be used as negative training examples for training the MLA.

traducir

MACHINE LEARNING MODEL ERROR DETECTION

Publication No.: US2021374601A1 02/12/2021

Applicant:

IBM [US]

Absstract of: US2021374601A1

A system includes a memory having instructions therein and at least one processor in communication with the memory. The at least one processor is configured to execute the instructions to determine a global-level importance magnitude value for a global-level importance of an explainable feature of a machine learning base model based on a first prediction of the machine learning base model. The at least one processor is also configured to execute the instructions to determine a global-level importance direction label for the global-level importance of the explainable feature based on the first prediction. The at least one processor is also configured to execute the instructions to generate a communication for presentation to a user based on a second prediction of the machine learning base model, based on the global-level importance magnitude value, and based on the global-level importance direction label.

traducir

Intelligent Dealership Recommendation Engine

Publication No.: US2021374616A1 02/12/2021

Applicant:

CAPITAL ONE SERVICES LLC [US]

US_11126931_B1

Absstract of: US2021374616A1

Aspects described herein may provide an interface and/or search functionality for a dealership to determine vehicles a customer is most likely to purchase. A recommender system may generate vehicle recommendations for a dealership to sell to a customer based on customer information, vehicle information, and dealership information. Machine learning may be used to generate the recommendations. The recommendations may be based on the vehicle preferences of a customer.

traducir

Performing Enhanced Exception Processing Using Cognitive Automation Tools

Publication No.: US2021374575A1 02/12/2021

Applicant:

BANK OF AMERICA [US]

Absstract of: US2021374575A1

Aspects of the disclosure relate to performing enhanced exception processing using cognitive automation tools. In some embodiments, a computing platform may receive interaction data identifying one or more actions performed by one or more users in resolving a plurality of exception items associated with an exception queue. Subsequently, the computing platform may train, using a learning engine, a machine learning model to resolve a first exception and a second exception of one or more exceptions based on the interaction data. Based on training the machine learning model, the computing platform may generate one or more configuration commands directing a processing module to implement the machine learning model to process additional exception items associated with the exception queue. The computing platform then may send the one or more configuration commands to the processing module.

traducir

ADAPTIVE MACHINE LEARNING SYSTEM FOR AN EDGE DEVICE

Publication No.: US2021374571A1 02/12/2021

Applicant:

ATOS INFORMATION TECH GMBH [DE]

EP_3916646_PA

Absstract of: US2021374571A1

An adaptive machine learning system (1) for an edge device comprises at least one sensor (2), at least one input compensation module (3) and an evaluation module (4). The sensor (2) is designed to capture input data. The input compensation module (3) is designed to modify the input data such that edge device specific artifacts of the input data are compensated. The evaluation module (4) is trained to process the modified input data and to generate output data as a result.

traducir

Crash Prediction System

Publication No.: US2021374567A1 02/12/2021

Applicant:

BANK OF AMERICA [US]

Absstract of: US2021374567A1

A crash prediction computing system includes a machine learning module capable of analyzing data logs associated with each of a plurality of services or applications to identify and categorize every error, exception, and/or crash, such as those resulting from client system interactions based on crash type, customer profile type, customer screen navigation flow, time or crash. The machine learning algorithms continuously train the crash prediction models for each crash category with associated client computing system navigation flow. The crash prediction computing system applies each model before each screen/activity navigation to predict whether the next move will result in an error, exception or crash, and for each predicted error, exception, or crash, automatically implement alternate route functionality to arrive at a desired target.

traducir

PREDICTIVE SCHEDULING AND EXECUTION OF DATA ANALYTICS APPLICATIONS BASED ON MACHINE LEARNING TECHNIQUES

Nº publicación: US2021374564A1 02/12/2021

Applicant:

CAPITAL ONE SERVICES LLC [US]

Absstract of: US2021374564A1

Disclosed herein are system, apparatus, article of manufacture, method, and/or computer program product embodiments for predictive scheduling and execution of data analytics applications based on machine learning techniques. An apparatus may operate by determining a first prediction entry in a predicted execution schedule based at least on a current timestamp. The apparatus may also operate by determining that a first confidence score of the first prediction entry is greater than or equal to a confidence score threshold and determining that an execution prediction of the first prediction entry is greater than or equal to an execution threshold. The apparatus may further operate by transmitting a first data analytics application execution request configured to request a first instance of execution of the data analytics application.

traducir

Page1/4 nextPage results/page

punteroimgGo back